Re: st: Using AIPW for missing data purposes in RCTs?

[Steve Samuels <sjsamuels@gmail.com>]

I agree with Ariel.  I don't know that any of the programs that have been mentioned,
-te aipw-, -dr-, or -drglm- , will solve Steve K.'s specific problem.


Steve

On Jun 27, 2013, at 10:53 AM, Ariel Linden, DrPH wrote:

Believe it or not, I am aware of the issues of missing data in both clinical
trials and observational studies. It was not clear from the original posting
what was the missing data problem. I did suggest -mi- if that suited the
problem. IPTW approaches to deal with missing data are another animal
altogether. Again, it was not clear from the original posting (nor is it any
more clearer now) what the problem is with the data. 

The reference I provided was for longitudinal data, where weighting is used
to account for censoring, which is a missing data problem.

The programs that Steve refers to below are not for longitudinal data.
Again, without knowing what your specific study is doing, or what missing
data you have, it is not a simple task of providing useful information. 

Ariel

________________________________________
From
  Steve Samuels <sjsamuels@gmail.com>
To
  statalist@hsphsun2.harvard.edu
Subject
  Re: st: Using AIPW for missing data purposes in RCTs?
Date
  Thu, 27 Jun 2013 09:54:03 -0400
________________________________________
You are welcome, Steve.

I overlooked -drglm-, "Doubly robust estimation in generalized linear
models" by N. Orsini, R. Bellocco, and A. Sjolander

See: http://www.stata-journal.com/article.html?article=st0290

. net install st0290.pkg

Also, -dr- is described in a Stata Journal article
http://www.stata-journal.com/article.html?article=st0149

Steve


On Jun 27, 2013, at 4:59 AM, Stephen Kay (TRUELIFE) wrote:

Thanks Steve for your clarifications to Ariel and also for the additional
references and ado file that I was not aware of - most useful. 

Here's a further reference for you Ariel - free to download and a really
good source to help in understanding various methods of handling missing
data in clinical trials:

http://www.nap.edu/catalog.php?record_id=12955

Best,

Steve 



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/